The present invention relates to a device for the absolute reading of an optical code carried by, more particularly, a disk with coded tracks, the reading taking place in this case by diascopy, that is to say by the differential reading of the transparent zones and of the opaque zones of a disk, disposed in a code known in the art under the name of Gray code.
It is known that such a device comprises--in combination with a disk comprising several concentric tracks each with a succession of fine alternating opaque zones and transparent zones--at least one light source adapted for illuminating the whole of the tracks and a coded reading assembly properly speaking which emits electric pulses representing finally the relative angular position of two elements of an assembly movable with respect to each other, the disk being carried by one of these elements.
Optical code reading devices are more and more widely used, one of their applications (to which the invention particularly applies) being that of determining the angular position of movable elements of a robot, by providing in this case a reading device for each movable element.
Although the invention applies particularly to the reading of a coded disk by diascopy, that is to say by transparency, it may also be applied to the reading of a coded disk by reflection and also to the reading of a coded strip by determining in this case not an angular position but a relative linear movement between two elements, one of which is movable in translation with respect to the other, the strip being carried by one of these two elements.
In what follows, the present description will be limited to the case of diascopy reading of a coded disk comprising several tracks using the Gray code.
Generally, a device for reading a coded disk comprises
on the one hand, at least one light source formed either by a filament lamp emitting light over a wide spectrum comprising the visible and the near infrared(the near infrared being the zone of maximum sensitivity for photoconductors, particularly silicon photoconductors), the filament lamp illuminating, through an appropriate optical system, the whole of the tracks of the disk, or an assembly of light emitting diodes emitting generally in the near infrared (in the 0.8 to 0.9 micron band),
on the other hand, a code reading assembly properly speaking, which cooperates with a coded disk and the light source or the light sources and which comprises first of all an array of photodetectors in number equal to the number of tracks, these photodetectors being formed more particularly by photodiodes or phototransistors which are either in the form of discrete components, or disposed in a common chip; in both cases, the number of photodetectors is equal to the number of light emitting diodes when the light source is formed by such diodes.
The photodetectors of the prior art deliver very weak signals and therefore they are followed by amplifiers, particularly by commercial operational amplifiers formed by integrated circuits with discrete components wired on a printed circuit, or else specific "prediffused" or hybrid circuits are used comprising in general as many amplifiers as there are coded tracks and so photoconductors (hereafter we will speak of reading channels, namely a channel for each track of the coded disk).
In each channel, the amplifier is followed usually by a shaping circuit generally formed by a trigger of the Schmitt trigger type outputting steep front pulses in response to each transparent zone of the track corresponding to the channel in question. Advantageously, in the same chip is disposed the Schmitt trigger assembly, and even more advantageously still in the same chip are disposed the amplifier asembly and the Schmitt trigger assembly either using the hybrid technique, or using the technique of prediffused circuits. There must further be provided for each Schmitt trigger, i.e. for each channel, an adjustment element for adjusting the switching threshold of the trigger so as to ensure, despite the variations in sensitivity of the photodetectors, correct operation of each channel and good switching accuracy of the subsequent electronic units.
In fact, it is necessary, for a large number of applications of code reading devices, to carry out logic processing before transmission of the information obtained for each channel by means of a photodetectoramplifier-Schmitt chain.
There are more particularly provided in general:
uncertainty removal means (in particular in the case of absolute multiturn coders or monoturn coders, when the disk is coded not with the Gray code, but with the natural binary code), these means being formed by commercially available integrated logic circuits;
means for transcoding the Gray code (which is the one appearing on the coded disk) into natural binary code (which is then the output code of the reading device) for facilitating subsequent use of the data from the reading device, these means being formed by commercially available "exclusive OR" circuits;
means for generating a parity bit for validating the transmission of a code when the number of bits is comformable (either even, or uneven, depending on the convention), particularly when the output of the reading device is a series output (after transformation of the parallel outputs of the Schmitt triggers into a series output ), these means being formed generally by an assembly of commercially available "exclusive OR" units;
means for transforming parallel outputs into a series output for reducing the number of connecting wires being the output of a reading device and the assembly using the position information, these means being formed by a commercially available shift register;
means for shifting the origin for replacing the fine mechanical setting of the disk by adjustment of the electric zero, these means being formed by coded switches.
The transmission of the information from the reading device is made conventionally
either by a series transmission, using as interface circuits line emitters and receivers which are formed by commercial chips,
or by parallel transmission of the information from the reading device over a "bus", then using a three state output (the high state not comprising any information output, whereas the other two states represent respectively the "0" bit and the "1" bit), or an output of the "open collector" type, for each channel, using commercially available circuits.
Finally, the power supply voltage is adjusted and filtered inside the reading device and the light source.
It is further necessary, in most cases, to compensate for the fluctuations or drifts of the single light source or of the light sources so as to ensure correct operation of the assembly over a wide temperature range.
The assembly comprising the power supply, its regulation and its filtration, on the one hand, and compensation of the fluctuations of the light source or sources, on the other, is formed conventionally by means of commercially available components.
It will be readily understood that the reading assembly which has just been described above requires, in the usual construction, a large number of components which, once wired on a printed circuit, occupy a considerable volume and are relatively expensive, while offering limited reliability.
To overcome these drawbacks, it has been proposed to provide hybridization of the electronic part (amplifiers, Schmitt triggers, logic means), or else to construct the electronic part in the form of several large functional blocks, constructing each of these blocks in the form of a prediffused circuit, or else combining hybridization and prediffused circuits. These three types of solutions reduce the space required and improve the reliability, but do not practically reduce the price of the assembly of the code reading device (light source and associated reading device properly speaking).
The present invention aims at simultaneously improving the three above mentioned factors (space required, reliability and cost) of the known reading devices, by reducing the space required, increasing the reliability and reducing the cost, while further allowing the assembly of the reader to be tested without increasing the complexity of the system.